While there is a long history of research identifying the risk factors for and consequences of youth violence, there has been somewhat less scientific study of the factors that protect youth from violent behavior, and the mechanisms through which protection continues. Identifying these modifiable protective factors and how they operate is important to develop policy that can reduce youth violence, injury, and death. The proposed research utilizes the Rochester Youth Development Study (RYDS, n=1000) and the Rochester Intergenerational Study (RIGS, n=500) to provide a comprehensive assessment of factors that protect youth from engaging in violent behaviors from childhood through adulthood. These studies follow three generations of families over 19 years and include a comprehensive set of violent outcomes, risk factors, and protective factors over much of the life course. The project has four specific aims. The first is to identify theoretically derived protective factors that statistically interact with risk factors, using quadratic equations, to decrease the likelihood that youth will participate in violent behavior. This will be done at each major stage of the lifecourse from childhood to adulthood. The second specific aim is to identify protective factors that can deflect individuals from various violent behavioral trajectories during adolescence and early adulthood. We will use semi-parametric growth curve models to accomplish this and we will use a novel technique to exchange protective resources between groups experimentally to determine if bolstering protective resources changes the individuals' violent offending [unreadable] trajectories. The third aim is to determine if, for specific histories of violent offending, the causal impact of protective factors is due to their unique effect or to other complex selection mechanisms. Here we will use propensity score matching emulating an experimental design to determine if the impact of the protective factor is real or a chimera of selection bias into the protective factor. The final specific aim is to study the causal impact of family and environmental protective factors predicting violent outcomes in young adults of one generation to assess the effect on their children. We will determine whether and how protective factors can benefit the generation to which they are applied as well as the children of that generation. [unreadable] [unreadable] [unreadable]